Juvenile rheumatoid arthritis (JRA) is the most common connective tissue disease in children, and the knee is the most frequently affected joint. The chronic arthritis of JRA can lead to cartilage destruction, bony erosions, crippling joint deformity and bony ankylosis. Early identification and treatment of JRA can improve long term functional outcome. Currently, plain radiography is used to assess anatomic disease progression. However, plain films reveal only late manifestations of the disease as described above. Magnetic resonance imaging (MRI) is an imaging modality that provides excellent soft tissue contrast and accurately depicts synovium, articular cartilage, ligaments, menisci and osseous structures in the knee. A longitudinal study of an inception cohort of children with JRA is proposed, using MRI as an imaging modality for the evaluation of disease severity ia the knee. In aim #1, the MRI findings of the knee in these patients will be identified. Fifty patients with newly- diagnosed JRA will undergo MRI of the knee on a 1.5 Tesla clinical magnet at the time of clinical presentation and at yearly intervals for up to 5 years. The variables that will be evaluated by MRI include extent of synovial proliferation, degree of synovial enhancement with gadolinium- DTPA, qualitative as well as quantitative volumetric assessment of cartilage, quantitative assessment of joint effusion, extent of osseous erosions and pannus invasion into bone, presence of physeal closure and degree of meniscal hypoplasia. The MR imaging results will be compared with plain films obtained at the same time. In aim #2, the MRI findings will be correlated with clinical assessment using a specific articular severity score for the imaged knee, and examiner global assessment of disease activity. In aim #3, the serial MRI and x-ray exams will allow time-oriented assessment of disease progression. In aim #4, MRI and clinical findings will be combined into regression models to determine which initial MRI variables best predict clinical, functional and radiologic outcomes. The long-term goals will be to determine the utility of MRI in monitoring patient response to therapies, and to develop an accurate and reproducible MRI scoring system for JRA. We believe MRI is a sensitive disease monitoring modality that will help guide the choice of therapeutic options, serve to predict clinical outcome and monitor patient response to therapy in JRA.